Outline

Motivation

Experimental PV installation spots

Figure 1: Experimental PV installation spots

Figure credit: EnBW.com, solarbusinesshub.com, Xinhua News Agency via Getty Images

PV in highway turnoffs: The idea

Scaling across Germany

Satellite imagery of two highway turnoffs in Brandenburg

Figure 2: Satellite imagery of two highway turnoffs in Brandenburg (:copyright: GeoBasis-DE/LGB, OpenStreetMap)

When is a turnoff suitable for PV?

Insights from pilot project “Lustnauer Ohren”

Derivation of the project’s objective

Conclusions for contextual focus of our project

Driving factors for one-off costs due to construction and operating costs:

Note: The environmental assessment determines one-off costs and long-term profitability of the PV.

Conclusions for our project from insights (pilot project)

Driving factors for one-off costs due to construction and operating costs:

Economic Model

The economic model used for the evaluation of turnoffs

Data acquisition

Selected challenges

Data preparation

Selected challenges

Figure 3: Example processing of a turnoff in Brandenburg (© GeoBasis-DE/LGB, OpenStreetMap)

Data preparation

Selected challenges

Figure 4: Example processing of a turnoff in Brandenburg (continued) (© GeoBasis-DE/LGB, OpenStreetMap)

Data analysis

Selected challenges

The height model of a selected driveway. The standard deviation of the slope is 4.68, the rating therefore 0.95 (© GeoBasis-DE/LGB)

Data analysis

Results

Training data - Postdam

First choice training dataset: Semantic Labeling Benchmark data for the German city Potsdam, provided by the International Society for Photogrammetry and Remote Sensing (ISPRS)

Label Color
1.) Impervious surfaces white
2.) Building blue
3.) Low vegetation light-blue
4.) Tree green
5.) Car yellow
6.) Clutter/background red

Data Source: 2D Semantic Labeling Contest - Potsdam. Available online: https://www.isprs.org/education/benchmarks/UrbanSemLab/default.aspx (accessed on 14 December 2022).

Training data - Postdam

Advantages

Disadvantages

Training data - Postdam

Color Labels: White - Impervious surfaces; Light-blue - Low vegetation; Green - Tree

Training data - China (regions of Nanjing, Changzhou and Wuhan)

Second choice training dataset: Land-cOVEr Domain Adaptive semantic segmentation (LoveDA) dataset

Label Color
1.) Background white
2.) Building blue
3.) Road light-blue
4.) Water green
5.) Barren yellow
6.) Forest red
7.) Agriculture red
0.) Unknown red

Training data - China (regions of Nanjing, Changzhou and Wuhan)

Advantages

Disadvantages

Land cover classification

Goal: Detect features in aerial images * Classes of interest + Agriculture + Buildings + Forest and Trees + Other

Land cover classification

Analysis tool: Image segmentation

Figure credit: Choi, U. (2021, March 19) Semantic Segmentation (FCN, U-Net, DeepLab V3+).

Semantic Segmentation - UNet

Model Advantages Disadv.
UNet - widely used basic segmentation model - training from scratch
- feedback from interim presentation - model overfits to training data
- good starting point - labelled data not specific for our task

Semantic Segmentation - ResNet50 + BigEarthNet (Sumbul et al., 2019)

Semantic Segmentation - ResNet50 + BigEarthNet (Sumbul et al., 2019)

Model Advantages Disadv.
ResNet50 - pre-trained weights - trained for classification task
+ BigEarthNet - seasonal satellite images - inferior predictions
- big dataset on European countries

Semantic Segmentation - ResNet101 + COCO (Lin et al., 2014)

Model Advantages Disadv.
ResNet101 - pre-trained weights - COCO dataset
+ COCO - segmentation task - trained for object segmentation
- superior predictions after few training iterations
- least overfitting
- complexity of model seems to fit task

Presentation of the result - Dashboard (Jan)

References